Technical Field
The present invention relates to a method and device for importing table format multi-dimensional data and translating the data into a format that can be easily processed by a computer.
Background Art
Nowadays, there is an increasing need for the following processes to be performed.
(1) A process of filtering multi-dimensional data, such as statistical data represented in a table format, from a variety of perspectives, and accumulating the resultant data.
(2) A process of integrating pieces of multi-dimensional data, which are represented in different table formats, into a single piece of multi-dimensional data.
However, multi-dimensional data represented in a table format is usually not suitable for computer processing. Thus, in recent years, multi-dimensional data has been attempted to be represented in a format called RDF (Resource Description Framework) so as to be adapted to the aforementioned computer processing. For example, when tabular data represented by three rows×three columns as shown in FIG. 1 or 2 is represented in the RDF format, it will be schematically represented in a network format as shown in FIG. 3. FIG. 3 represents “cell 1,” which is contained in the data portion (i.e., 2 rows×2 columns of cells) of the table shown in FIGS. 1 and 2, in the RDF format. Data in each cell is described with dimension information and measure information for uniquely identifying the cell. When such a description method is adopted, even if the same cell is described in different table formats, the cell will be represented the same in the RDF format, whereby it becomes easier to perform a filtering process, an integration process, or the like with a computer.
As described above, although table formats are usually not suitable for computer processing, they are easy to handle by humans. Meanwhile, when thousands of pieces of multi-dimensional data are represented in the RDF format, problems would arise that humans would not be able to easily recognize the relationship among the pieces of data and that the RDF format lacks a data listing property. In addition, another problem would arise that with the RDF format, it would be difficult for humans to find out the intended data and further edit the data. Further, while a table format allows elements of row headings or column headings (e.g., “shipment value of products” in FIG. 2) to be changed at once, if the same process is to be executed on data in the RDF format, it would be necessary to change all pieces of information on the corresponding nodes and links (e.g., the node of “ex:vStocks” and the link thereof in the case of FIG. 3) for individual cells.